CN114759959B - Phased array beam forming method for inhibiting interference between beams - Google Patents

Abstract

The invention discloses a phased array beam forming method for inhibiting interference among beams, which takes the linear combination of the power of the minimum beam in the interference direction and the power of the beam in the central area of a main lobe as an optimization target by introducing a weight factor under the condition of meeting the constant mode constraint of a phased array, and optimally designs single or multiple beams by means of a manifold optimization tool box. In the millimeter wave multiuser communication system, the invention can achieve the average and rate performance of approximate mixed beam forming under the condition of using only analog beam forming and not using digital beam forming, and has effective inhibition effect on interference among users. In a radar system, the phased array beam forming method provided by the invention has better interference suppression performance compared with the existing method.

Description

A phased array beamforming method to suppress inter-beam interference

Technical field

The present invention relates to the technical field of phased array beamforming for suppressing inter-beam interference, and in particular, to a phased array beamforming method for suppressing inter-beam interference.

Background technique

In recent years, with the development of wireless communications, wireless phone services have been continuously upgraded. Mobile communication systems are facing explosive growth in data traffic and massive device connections, which has stimulated research on the application of millimeter waves in wireless communications. The wavelength of millimeter waves is short, only a few millimeters, which enables millimeter wave communications to be carried out with smaller antenna sizes. The transceiver end can configure a large-scale antenna array in a limited area. Millimeter wave systems mainly use this large-scale antenna Arrays form beams to enable high-rate transmission. However, due to the existence of a large number of antennas, assigning a dedicated RF link to each antenna is not only difficult to implement in hardware, but also causes high power consumption and high RF link costs. Therefore, using a small number of RF links to pass a certain number of mobile Hybrid architectures in which phase drivers drive all antenna elements are widely used in current millimeter-wave systems.

In a millimeter-wave multi-user wireless communication system, different data streams can occupy the same time-frequency resources and form beams to transmit different data through space division multiplexing. In actual projects, there will be mutual interference between different data streams. Constrained by the number of radio frequency chains, it is necessary to design the beam for transmitting data through a phased array network composed of phase shifters to suppress interference between data streams. However, since the phase shifter only has an adjustable phase, each element in the phased array has a constant modulus constraint. Therefore, how to perform beamforming design on the phased array to suppress inter-beam interference is a key issue in millimeter-wave multi-user wireless communication systems. The hard part.

In addition, in radar systems, it is usually desirable to suppress the interference of clutter and other signals by minimizing the power of signals transmitted to and received from the noise source. Therefore, the interference suppression design of the beam is useful for improving the performance of the radar system. Significance. Forming a null in a specific direction of the beam pattern is an effective anti-interference technology in radar systems. Pure phased array networks have attracted much attention in large phased array systems due to their economy and simplicity of the feed network. How to design the beam to form a null in a specific direction to suppress interference is also a difficulty in radar systems with constant modulus constraints.

Contents of the invention

In view of this, the object of the present invention is to provide a phased array beamforming method that suppresses inter-beam interference. This method minimizes the interference of the beam in the interference direction by introducing a weighting factor while satisfying the constant mode constraint of the phased array. The linear combination of power and beam power in the main lobe center area is the optimization goal. With the help of the manifold optimization toolbox, single or multiple beams are optimally designed.

In order to achieve the above objects, the present invention adopts the following technical solutions:

A phased array beamforming method for suppressing inter-beam interference, the method includes the following steps:

Step S1. For a wireless system, construct an optimization problem with the optimization goal of minimizing the linear combination of the power of the beam in the interference direction and the power of the beam in the main lobe center area; wherein the wireless system includes a millimeter-wave multi-user wireless communication system. and radar systems, the described optimization problem satisfies the constant modulus constraints of all phase shifters in the phased array;

Step S2: Design single or multiple beams for the wireless system by solving the optimization problem constructed in step S1.

Further, in step S1, when the wireless system is a millimeter wave multi-user wireless communication system, its downlink signal transmission model is first constructed to minimize the power of the current user beam in its main lobe center area and the power of this beam in the rest of the The linear combination of the power in the center area of the main lobe of the user beam constructs an optimization problem for the optimization objective.

Further, in step S1, if the base station can obtain the accurate departure angle of each user, the main lobe center area is simplified to a center point.

Further, the downlink signal transmission model is specifically expressed as:

Among them, y _q represents the signal received by the user; Represents the downlink channel vector between the base station and the q-th user,/> Represents the complex number field, (·) ^H represents the conjugate transpose operation;/> K column vectors/> Represents the beamforming vector sent by the base station to K users;/> Represents the data stream sent by the base station to K users, which satisfies _{The power} constraint _of A complex Gaussian distribution with 0 and variance σ ² .

Further, in the downlink signal transmission model, the channel is modeled as:

Among them, N _t represents the number of base station antennas, L _q represents the total number of multipaths in the channel between the base station and the q-th user, α _q,l and φ _q,l represent the complex gain and departure value of the l-th path of the q-th user respectively. angle; a(φ _q,l ) represents the array guidance vector of the l-th path of the q-th user, and its specific expression is:

Among them, λ _c is the carrier wavelength, d represents the antenna array element spacing, and (·) ^T represents the transposition operation.

Further, the optimization problem is specifically expressed as:

Among them, λ is a positive real weight factor, θ _k,j represents the j-th sampling point in the main lobe center area of the k-th beam, k=1,2,…,K, j=1,2,…, J; w _q represents the qth column of F _RF ; Represents the total power of the sampling points of the qth beam in the main lobe center area of other beams, which is defined as the sum of the interference power of the qth beam to the other beams; M represents the number of sampling points in the main lobe center area of the qth beam, / > Represents the sum of power of the M sampling points of the q-th beam in the center area of its main lobe; |·| represents the modulus value, [w _q ] _n represents the n-th element of the beamforming vector w _q , according to all phases of the phased array The constant modulus constraint of the phase shifter, each element in w _q satisfies

Furthermore, when the base station can obtain the precise departure angle of each user and the main lobe center area is simplified to a center point, the corresponding optimization problem is specifically expressed as:

Among them, θ _k represents the main lobe center point of the k-th beam.

Further, when the wireless system is a radar system, an optimization problem is constructed with the optimization goal of minimizing the linear combination of the power of the beam in its main lobe center area and the power of the beam in the interference area.

Further, when the wireless system is a radar system, the optimization problem is specifically expressed as:

Among them, λ is a positive real weight factor, w represents the beamforming vector, θ _i,j represents the j-th sampling point in the i-th interference area, Represents the total power of the beam's sampling points in the interference area; M represents the number of sampling points in the main lobe center area of the beam,/> Represents the sum of power of the M sampling points of the beam in the center area of its main lobe; [w] _n represents the nth element of the beamforming vector w. According to the constant modulus constraints of all phase shifters of the phased array, each of w All elements are satisfied/>

Further, in step S2, the optimization problem is solved through the manifold optimization toolbox.

The beneficial effects of the present invention are:

1. Regarding the millimeter wave multi-user beamforming problem under the fully connected architecture, the present invention establishes an optimization problem model by considering the interference suppression between the beams transmitted by the base station to different users. Based on this, a beamforming method is proposed, which can only Achieve average and rate performance that approximates hybrid beamforming using analog beamforming instead of digital beamforming;

2. Regarding the problem of phase-only beam design to suppress interference in radar systems, the present invention provides an effective beam design scheme, which has better performance than the existing phase-only beam nulling design scheme.

Description of the drawings

Figures 1 and 2 are schematic diagrams of a millimeter wave multi-user wireless communication system model used in Embodiment 1 of the present invention;

Figure 3 shows that the base station is equipped with 64 array elements, the array element spacing is half a wavelength, the number of radio frequency links is 4, the base station obtains an accurate departure angle, and the total number of transmission paths between the user and the base station is equal to 1, using the present invention Comparison diagram of the average user sum rate of the beamforming method provided in Embodiment 1 and the average user sum rate of hybrid beamforming and fully digital beamforming;

Figure 4 shows that when the base station is equipped with 64 array elements, the array element spacing is half a wavelength, the number of radio frequency links is 4, and the total number of transmission paths between the user and the base station is equal to 1, Embodiment 1 of the present invention is used under imprecise departure angle conditions. Comparison diagram of the user average sum rate of the beamforming method designed under the conditions of hybrid beamforming and full digital beamforming designed under precise departure angle conditions;

Figure 5 shows that when the base station is equipped with 64 array elements, the array element spacing is half a wavelength, the number of radio frequency links is 4, and the total number of transmission paths between users and the base station is equal to 3, Embodiment 1 of the present invention is used under imprecise departure angle conditions. Comparison diagram of the user average sum rate of the beamforming method designed under the conditions of hybrid beamforming and full digital beamforming designed under precise departure angle conditions;

Figure 6 shows a radar antenna array equipped with 32 array elements, and the array element spacing is half a wavelength, using the beam designed in Embodiment 2 of the present invention and using the semidefinite relaxation algorithm in the literature [1] and the Cronet in the literature [2] Comparison of beams designed by g-decomposition algorithm.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

Example 1

Referring to Figures 1 to 5, this embodiment provides a phased array beam method for suppressing inter-beam interference for a millimeter wave multi-user wireless communication system. The method specifically includes the following steps:

Step S1: Construct a millimeter-wave multi-user wireless communication system model. The structure of the model is shown in Figure 1. Specifically:

For a downlink communication scenario where a base station serves K users, all of which are single-antenna users, the base station adopts a fully connected hybrid beamforming architecture, the number of radio frequency links is equal to the number of users, that is, N _RF =K, and the antenna array has N _t A uniform linear array with antennas and antenna spacing of half a wavelength.

Specifically, taking the q-th user as an example, its downlink signal transmission model can be established as:

Among them, y _q represents the signal received by the user; Represents the downlink channel vector between the base station and the q-th user,/> Represents a complex number field;/> Represents the data stream sent by the base station to K users, which satisfies/> _{The power} constraint _of is 0 and is a complex Gaussian distribution with variance σ ² , that is,/> Represents the simulated beamforming matrix of the base station. The simulated beamforming is performed on a phased array network composed of phase shifters, and is constrained by the constant modulus value of the phase shifter, that is, each element in F _RF satisfies/> Represents the digital beamforming matrix of the base station.

Specifically, in this embodiment, the beamforming method used is phased array beamforming instead of digital beamforming. Therefore, the digital beamforming matrix is set to the unit matrix, that is, F _BB =I _K , then figure The hybrid beamforming architecture in Figure 1 is simplified to the analog beamforming architecture in Figure 2. The above downlink signal transmission model is simplified to:

Among them, the K column vectors of F _RF Represents the beamforming vectors sent by the base station to K users respectively.

Specifically, in this embodiment, in the above downlink signal transmission model, the channel is modeled as:

Among them, L _q represents the total number of multipaths in the channel between the base station and the q-th user, α _q,l and φ _q,l respectively represent the complex gain and angle of departure (AOD) of the l-th path of the q-th user. ). a(φ _q,l ) represents the array guidance vector of the l-th path of the q-th user, and its specific expression is:

where λ _c is the carrier wavelength, d represents the antenna array element spacing, and (·) ^T represents the transposition operation.

Step S2: For the millimeter-wave multi-user wireless communication system model constructed in step S1, establish a beamforming model with constant mode constraints that suppresses inter-beam interference;

Specifically, since the users are all single-antenna single-RF chain users, in order to maximize the signal power received by the users, the main lobe centers of the K beams formed by the simulated beamforming matrix F _RF need to be aligned with the maximum complex gain of the K users respectively. The AOD of the path (usually a line-of-sight path) is recorded as However, due to the existence of side lobes, interference inevitably exists between beams, which affects the communication between the base station and the user.

Step S201: Depending on whether the base station can obtain accurate information for each user To construct different optimization problems, if the base station cannot obtain accurate information for each user/> Then, an optimization problem is constructed with the optimization goal of minimizing the linear combination of the power of the current user beam in the center area of its main lobe and the power of the beam in the center area of the main lobe of other user beams.

Specifically, in this embodiment, step S201 includes:

In order to reduce the impact of inter-beam interference on communication quality, it is necessary to suppress the inter-beam interference. In addition, considering that the base station may be affected by the beam resolution during the beam scanning stage, it cannot obtain accurate information for each user. And can only get/> Located in an angular domain range Ω _k with a width of beam resolution, there is a pair/> The positioning error is the beam scanning error, so when the base station actually transmits the beam, the beam center is aligned with the center of the angular domain range Ω _k , rather than the precise/> precise/> may be distributed anywhere in Ω _k . Therefore, in order to suppress interference between beams transmitted to different users while ensuring the communication quality of each user, it is necessary to make a single beam form a null in the Ω _k region of the remaining beams under the condition that the main lobe offset is as small as possible. The design problem of beams for multiple users is K sub-problems that are independent of each other.

More specifically, in this embodiment, taking the design of a beam transmitted to the q-th user as an example, the design problem of the beamforming vector w _q =F _RF (:,q) can be established as the optimization of the null notch in the beam area question:

Among them, λ is a positive real weight factor, θ _k,j represents the j-th sampling point in the main lobe center area of the k-th beam, k=1,2,…,K, j=1,2,…, J; w _q represents the qth column of F _RF ; Indicates the total power of the sampling points where the beam of the qth user is allocated in the main lobe center area Ω _k of other users' beams, that is, the total interference power caused by the beam of the qth user to the beams of other users; M represents the current beam to be designed That is, the number of sampling points in the main lobe center area of the q-th beam,/> Represents the sum of power of the M sampling points of the q-th beam in the center area of its main lobe; |·| represents the modulus value, [w _q ] _n represents the n-th element of the beamforming vector w _q , according to all phases of the phased array The constant modulus constraint of the phase shifter, each element in w _q satisfies/>

Step S202: If the base station can obtain accurate information of each user Then there is no need to sample the main lobe center area Ω _k of each beam. The main lobe center area in the above optimization problem is simplified to a center point. Therefore, the optimization problem of the beam area null constructed in step S201 can be simplified to multiple The optimization problem of point zero depression is specifically expressed as:

Among them, θ _k represents the main lobe center point of the k-th beam.

Step S203: Solve the optimization problem of the beam area nulling or the optimization problem of multi-point nulling constructed above, which includes: setting the weight factor λ, and solving the optimization problem through the manifold optimization toolbox to obtain the beam forming Vector w _q . By analogy, w _k (1≤k≤K) can be solved in sequence, and the simulated beamforming matrix F _RF can be obtained, where F _RF (:,k)=w _k .

Example 2

Referring to Figure 6, this embodiment provides a phased array beam design method for suppressing inter-beam interference in a radar system. This method is based on a radar antenna array with an element spacing of half a wavelength. The number of array elements is N _t . Arranged in a uniform linear array. The method specifically includes the following steps:

Step S1: Determine the radar system and construct its beam pattern expression, specifically including:

Step S101. For the uniform linear array, its array steering vector in the θ direction is expressed as:

Step S102. Assume the beamforming vector is w, then its corresponding beam pattern can be expressed as:

f(θ)＝|w ^H a(θ)| ²

Among them, |·| represents the modulus value.

Step S2: Design the beam pattern constructed in step S1. The design includes: minimizing the linear combination of the power of the beam in its main lobe center area and the power of the beam in the interference area as the optimization goal. problem, and then solve the optimization problem to obtain the beamforming vector.

Specifically, in this embodiment, step S2 specifically includes:

Step S201. In order to suppress interference in a specific direction, the beam needs to be designed so that the beam pattern forms a null in a specific direction. This problem can be transformed into minimizing the power allocated to the beam in a specific direction. At the same time, in order to minimize the offset of the main lobe, it is also necessary to ensure that the power of the beam is maximized in the center area of the main lobe. The design problem of the beamforming vector w can be established as the following optimization problem:

Among them, λ is a positive real weight factor, θ _i,j represents the j-th sampling point in the i-th interference area, i=1,2,...,I, Indicates the total power of the beam at the sampling points in the interference area. M represents the number of sampling points in the center area of the main lobe of the beam,/> Represents the total power of M sample points of the beam in the center area of its main lobe. |·| represents the modulus value, [w] _n represents the nth element of the beamforming vector w. According to the constant modulus constraints of all phase shifters in the phased array, each element in w satisfies/>

Step S202: Set the weight factor λ, and solve the optimization problem through the manifold optimization toolbox to obtain the beamforming vector w.

In order to verify the correctness and advancement of the methods in Example 1 and Example 2 above, simulation experiments were conducted, specifically including:

The simulation parameters in Figure 3 are: the number of base station antennas N _t is 64, the number of radio frequency links N _RF is 4, and the base station serves a total of K = 4 users. The total number of transmission paths L _k between the user and the base station is equal to 1, including only 1 main path, and the main path gain obeys the complex Gaussian distribution, that is The channel state information at the base station is the precise path AOD, and the weight factor λ=1000. In Figure 3, step S202 in Embodiment 1 is first used to establish an optimization problem model, and then the beamforming vector of each user is obtained through the manifold optimization toolbox to form a simulated beamforming matrix. Combined with the actual channel, the transmission signal-to-noise ratio was changed, 2000 Monte Carlo simulations were performed, and the relationship curve between the user's average sum rate and SNR was drawn, as shown by the circular solid line in Figure 3. At the same time, draw the relationship curve between the user average sum rate and SNR under hybrid beamforming conditions, as shown by the asterisk solid line in Figure 3, and the relationship curve between the user average sum rate and SNR under fully digital beamforming conditions, as shown Shown by the dotted line in Figure 3. The expression for the average sum rate of users is

Among them, R _k represents the reachable rate of the k-th user, and its specific expression is

Among them, F=F _RF F _BB , and F for hybrid beamforming, full digital beamforming and phased array beamforming proposed by the present invention are respectively F ₂ , F ₃ and F ₁ . F ₁ =F _RF for phased array beamforming; hybrid beamforming is the asterisk solid line It is obtained by performing zero-forcing precoding on the equivalent channel matrix _Heq = HF _RF , that is, F _BB = ( _Heq ) ^-1 = (HF _RF ) ^-1 , where H = [h ₁ ,h ₂ ,...,h _K ] ^H , preliminary hybrid beamforming matrix F _HB =F _RF F _BB . Since hybrid beamforming does not have power gain, it is necessary to perform energy normalization processing on each column of F _HB to obtain the final F ₂ , that is /> Fully digital beamforming/> For a zero-forcing precoder, F ₃ = ^HH (HH ^H ) ^-1 . Comparing the above three curves, it can be found that the beamforming method proposed by the present invention can achieve the same performance as hybrid beamforming, and compared with all-digital beamforming, the average user sum rate of the two only differs by 0.0457bps when the SNR is 15dB. /Hz. This is because the phased array beamforming method proposed by the present invention designs the user's beam so that the energy allocated to the current user's beam at the center position of other users' beams is as close to zero as possible, and the equivalent channel matrix _Heq = HF _RF The modulus value of the non-diagonal elements is close to 0, which completes the task of inter-user interference suppression, so it can achieve the same performance as hybrid beamforming.

The simulation parameters in Figure 4 are: the number of base station antennas N _t is 64, the number of radio frequency links N _RF is 4, and the base station serves a total of K = 4 users. The total number of transmission paths L _k between the user and the base station is equal to 1, and the path gain of the main path obeys the complex Gaussian distribution, that is The channel state information at the base station is an inaccurate path departure angle, and the weight factor λ=1000. In Figure 4, step S201 in Embodiment 1 is first used to establish an optimization problem model, and then the beamforming vector of each user is obtained through the manifold optimization toolbox to form a phased array beamforming matrix. Combined with the actual channel, the transmission signal-to-noise ratio is changed, and 2000 Monte Carlo simulations are performed to draw the relationship curve between the user's average sum rate and SNR, as shown by the circular solid line in Figure 4. At the same time, draw the relationship curve between the user average sum rate and SNR under hybrid beamforming conditions, as shown by the asterisk solid line in Figure 4, and the relationship curve between the user average sum rate and SNR under fully digital beamforming conditions, as shown Shown by the dotted line in Figure 4. Among them, in order to indicate the performance upper bound, H used when calculating the hybrid beamforming matrix and the fully digital beamforming matrix is the precise channel state information, that is, the precise path departure angle. Comparing the above three curves, it can be found that the beamforming method proposed by the present invention can achieve the same performance as hybrid beamforming when SNR is less than 20dB, and compared with full digital beamforming, the average user rate of the two is when SNR is 15dB. The difference is only 1.13bps/Hz. This is because the phased array beamforming method proposed by the present invention designs the user's beam so that the energy allocated to the current user's beam in the center area of other users' beams is as small as possible, and the equivalent channel matrix _Heq = HF _RF is close to The diagonal matrix plays the role of interference suppression, so it can achieve the same performance as hybrid beamforming when the SNR is small, that is, when noise is the main factor affecting the sum rate.

The simulation parameters in Figure 5 are: the number of base station antennas N _t is 64, the number of radio frequency links N _RF is 4, and the base station serves a total of K = 4 users. The total number of transmission paths L _k between the user and the base station is equal to 3, including 1 main path and 2 slave paths. The path gain of the main path obeys the complex Gaussian distribution, that is The path gain of the slave path also obeys the complex Gaussian distribution, and the energy is 1/100 of the main path, that is/> The channel state information at the base station is an inaccurate path departure angle, and the weight factor λ=1000. In Figure 5, step S201 in Embodiment 1 is first used to establish an optimization problem model, and the beamforming vector of each user is obtained through the manifold optimization toolbox to form a simulated beamforming matrix. Combined with the actual channel, the transmission signal-to-noise ratio was changed, 2000 Monte Carlo simulations were performed, and the relationship curve between the user's average sum rate and SNR was drawn, as shown by the circular solid line in Figure 5. At the same time, draw the relationship curve between the user average sum rate and SNR under hybrid beamforming conditions, as shown by the asterisk solid line in Figure 5, and the relationship curve between the user average sum rate and SNR under fully digital beamforming conditions, as shown Shown by the dotted line in Figure 5. Among them, in order to indicate the performance upper bound, H used when calculating the hybrid beamforming matrix and the fully digital beamforming matrix is the precise channel state information, that is, the precise path departure angle. Comparing the above three curves, it can be found that due to the interference from the two slave paths, the performance of the phased array beamforming proposed by the present invention is lower than the performance of the hybrid beamforming combined with zero-forcing precoding, but it can still be achieved when SNR<10dB Achieve the same performance as hybrid beamforming. This is because the phased array beamforming method proposed by the present invention, when designing the user's beam, makes the energy allocated by the current user's beam in the center area of other users' beams as small as possible, which plays the role of interference suppression.

The simulation parameters in Figure 6 are: the number of antennas in the radar uniform linear array N _t =32, and the array element spacing is half a wavelength. The beam center angle θ ₀ =0°, the interval corresponding to the main lobe center area is [-1.7°, 1.7°], and the number of sampling points M=5. The interference area, that is, the null interval is [-10.8°, -14.45°] ∪ [14.9°, 18.2°], that is, I = 2. The number of sampling points in each interference area is J = 10, and the weight factor λ = 5000. In Figure 6, step S201 in Embodiment 2 is first used to establish an optimization problem model, the beamforming vector w is solved through the manifold optimization toolbox, and the beam corresponding to w is drawn, as shown by the solid line in Figure 6. At the same time, draw the beam designed by the semidefinite relaxation algorithm in literature [1], the beam designed by the Kronecker decomposition algorithm in literature [2], and the quasi-static beam without beam null design, as shown in the dotted line in Figure 6 Shown as lines, dotted lines and dashed lines. Comparing the above four curves, it can be found that in the null interval, the array gain of the beam designed in the present invention reaches below -60dB, which is lower than the array gain of literature [1] and literature [2], and is different from the quasi-static beam in this range. The array gain peak value in the interval has dropped by about 40dB; at the beam center position, compared with the quasi-static beam, the array gain of the beam designed in the present invention has only dropped by 1.16dB. Compared with the beam designed in the literature [1], the array gain The difference is only 0.26dB, and compared with the beam designed in literature [2], the array gain is 10.06dB higher. It can be seen from this that the beam designed in the present invention can achieve optimal interference suppression performance on the premise of ensuring the array gain in the main lobe center region.

The above-mentioned documents [1] are: P.J.Kajenski.Phase Only Antenna Pattern Notching Viaa Semidefinite Programming Relaxation[J]. IEEE Transactions on Antennas andPropagation, 2012, 60(5):2562-2565.

The above-mentioned documents [2] are: Gu T, Zhang X, He Z, et al. Phase-Only Nulling for Uniform Linear Array via Kronecker Decomposition [A]. In: 2021 GASS)[C].2021,pp.1-4.

In summary, the phased array beamforming method for suppressing inter-beam interference provided by the present invention can achieve the average sum rate performance of approximate hybrid beamforming under the condition that only analog beamforming is used instead of digital beamforming. It effectively suppresses interference between user beams. At the same time, when designing a single beam in a radar system, the beam null design scheme provided by the present invention has better performance than the existing method.

Everything that is not described in detail in the present invention is a well-known technology for those skilled in the art.

The preferred embodiments of the present invention are described in detail above. It should be understood that those skilled in the art can make many modifications and changes based on the concept of the present invention without creative efforts. Therefore, any technical solutions that can be obtained by those skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention and on the basis of the prior art should be within the scope of protection determined by the claims.

Claims (4)

1. A phased array beamforming method for suppressing inter-beam interference, the method comprising the steps of:

step S1, aiming at a wireless system, constructing an optimization problem with the linear combination of the power of the minimized wave beam in the interference direction and the power of the wave beam in the central area of the main lobe as an optimization target; the wireless system comprises a millimeter wave multiuser wireless communication system and a radar system, and the optimization problem satisfies the constant mode constraint of all phase shifters in the phased array;

step S2, designing a single beam or a plurality of beams for the wireless system by solving the optimization problem constructed in the step S1;

in the step S1, when the wireless system is a millimeter wave multi-user wireless communication system, firstly, constructing a downlink signal transmission model thereof, and constructing an optimization problem by taking a linear combination of the power of a current user beam in a main lobe central area thereof and the power of the beam in main lobe central areas of other user beams as an optimization target;

the downlink signal transmission model specifically comprises the following steps:

wherein y is _q A signal representing the user's receipt;representing the downlink channel vector between the base station and the qth user, for example>Representing the complex field, (. Cndot. ^H Representing a conjugate transpose operation; />K column vectors +.>A beamforming vector representing the transmission of the base station to K users; />Representing the data streams sent by the base station to K users, which satisfy +.>Power constraint, P _s The q-th element s representing the transmission power of the base station s] _q Data representing the transmission of the base station to the qth user; η (eta) _q Representing the additive white noise received by the qth user, subject to a mean of 0 and a variance of sigma ² Complex gaussian distribution of (a);

in the downlink signal transmission model, the channel modeling is as follows:

wherein N is _t Indicating the number of base station antennas L _q Representing the total number of multipaths, alpha, of the channel between the base station and the qth user _q,l And phi _q,l The complex gain and departure angle of the first path of the q-th user are respectively represented; a (phi) _q,l ) An array steering vector representing the first path of the qth user, the specific expression of which is:

wherein lambda is _c For carrier wavelength, d represents the antenna element spacing, (·) ^T Representing a transpose operation;

the optimization problem is specifically expressed as:

wherein λ is a positive real weight factor, θ _k,j A J-th sampling point, k=1, 2, …, K, j=1, 2, …, J, representing the central region of the main lobe of the kth beam; w (w) _q Represents F _RF Is the q-th column of (2);the power sum of the sampling points of the q-th beam in the central area of the main lobe of other beams is defined as the interference power sum of the q-th beam to the other beams; m represents the number of sampling points in the center region of the main lobe of the qth beam, +.>Representing the sum of the power of the M sampling points of the qth beam in the central region of its main lobeAnd; i·| represents modulo value, [ w ] _q ] _n Representing beamforming vector w _q According to the constant modulus constraint of all phase shifters of the phased array, w _q Each element of (a) satisfies

When the wireless system is a radar system, constructing an optimization problem for an optimization target by using a linear combination of the power of the minimum beam in the central area of a main lobe of the wireless system and the power of the beam in an interference area;

when the wireless system is a radar system, the optimization problem is specifically expressed as:

wherein λ is a positive real weight factor, w represents a beamforming vector, θ _i,j The J-th sample point at the I-th interference region, i=1, 2, …, I, j=1, 2, …, J;representing the sum of the powers of the beams at the sampling points of the interference area; m represents the number of sampling points in the central region of the main lobe of the beam,/->Representing the sum of the powers of the M sampling points of the beam in the central region of the main lobe of the beam; [ w ]] _n An nth element representing a beamforming vector w, each element in w satisfying +.>

2. The phased array beamforming method according to claim 1, wherein in said step S1, if the base station can obtain an accurate departure angle for each user, the main lobe center area is reduced to a center point.

3. The phased array beamforming method for suppressing interference between beams according to claim 1, wherein when the base station can obtain an accurate departure angle of each user, the main lobe central area is simplified to a central point, the corresponding optimization problem is specifically expressed as:

wherein θ _k Representing the main lobe center point of the kth beam.

4. A phased array beamforming method for suppressing inter-beam interference according to any of claims 1-3, wherein in said step S2, the optimization problem is solved by a manifold optimization tool box.